The Effect of Ketamine on Interleukin-6 Synthesis in Hepatic Resections Requiring Temporary Porto-arterial Occlusion

Introduction: Many complications in liver resection surgery, such as post-operative hepatic insufficiency, sepsis, and multi-organ liver failure, have been attributed to ischemia-reperfusion injury and the release of pro-inflammatory agents. Previous studies had shown that ketamine inhibited the synthesis of interleukin 6 (IL-6) in some cardiac surgeries, and that IL-6 levels had a direct correlation with mortality in critically ill patients.

Goal: The purpose of our study was to determine the effect of ketamine on IL-6 levels in patients undergoing liver resections with temporary porto-arterial occlusion (Pringle maneuver).

Material and Methods: Our prospective, controlled, randomized, and blinded study was approved by the Research Ethics Committee of the Hospital Italiano of Buenos Aires (CEPI). All procedures were carried out according to the Declaration of Helsinski. All patients voluntarily consented to the study and signed the appropriate informed consent approved by the CEPI.

Inclusion criteria were age of 21 years or older and planned liver resection with Pringle maneuver lasting 30-60 minutes. Those with chronic illness requiring corticosteroids, cirrhosis, hemodynamic instability prior to the surgery, diabetes, sepsis, surgical interventions or chemotherapy treatments within the past 30 days, pregnancy, illnesses that could potentially affect the hepatic circulation, arterial or ocular hypertension (contraindications for the use of ketamine), ketamine allergy, preoperative portal embolization / radiofrequency ablation, or requiring emergency surgeries were not considered. From March 2002 to June 2008, 44 consecutive patients agreed to participate and were enrolled in the study. Those who did not require the Pringle maneuver during the resection, who did not undergo the planned procedure, or whose hematocrit was less than 20% for over 30 minutes were excluded.

Patients were assigned to one of two groups according to a computer generated randomization.

The study group received ketamine 0.25 mg/kg, while the control group was administered an identical volume of saline. Syringes containing 10 ml of either ketamine or saline were delivered by the hospital pharmacy personnel to the corresponding anesthesiologist, who was blinded to their contents. In order to calculate the correct dose that was administered immediately after induction of anesthesia, the ketamine and placebo concentration was established at 10 mg/ml. Both anesthesia and surgical teams were the same in all cases. Members of both teams, as well as all personnel involved with blood collection remained blinded at all times.

All patients were transported to the operating room with an intravenous line in place and premedicated with midazolam 0.04 mg/Kg. Once in the operating room, they received intravenous antibiotics, non-invasive cardiac monitoring, blood pressure monitoring, and pulse oxymetry.

Remifentanil 0.25 micrograms/kg/min was administered prior to induction with sodium thiopental 2-2.5 mg/Kg. Vecuronium 0.1 mg/Kg was used for muscle relaxation. After waiting approximately 3 minutes, patients were intubated endotracheally and a nasogastric tube was placed. Remifentanil 0.5 micrograms/kg/min, ibuprofen 10 mg/Kg, and morphine 0.15 mg/kg were administered during the case.

Anesthesia was maintained with Sevorane in the setting of an FiO2 of 0.70. Mechanical ventilation was adjusted to allow an EtCO2 of 25-30 mmHg and a plateau pressure \< 30 cmH2O. An arterial line was placed after induction of anesthesia to serve for invasive monitoring as well as for blood sampling. A central line was placed in the right internal jugular vein to monitor intra-operative central venous pressure (CVP). All patients received body warmers and warmed fluids. We aimed for a CVP \< 5 cm H2O at the time of resection to diminish bleeding. Intravenous fursemide at doses of 10-20 mg as well as fluid restriction were employed when necessary in order to reach the desired value. Potassium levels were kept at or above 3.5 mmol/lt. Phenylephrine was used when necessary to maintain a median arterial pressure of at least 70 mmHg.

Patients who underwent extensive resections were admitted to the Intensive Care Unit (ICU) and maintained on mechanical ventilation for 6-8 hours prior to extubation. In all other cases patients were extubated at the end of the procedure, observed in the post anesthesia care unit (PACU) for at least 8 hours, and subsequently transferred to the floor if hemodynamically stable. Pain management in patients extubated intraoperatively was with synthetic opioids (dextropropoxifen 1 mg/kg and dipyrone 2.5 mg. In cases of persistent pain (4 or more in a visual scale of 10), analgesia was supplemented with 2 mg of morphine every 20 minutes until relief of symptoms, somnolence, or a respiratory rate of 8 or less per minute was observed.

Blood samples for IL-6 levels were obtained prior to surgery upon placement of the first intravenous line, and at 4, 12, 24, 72 and 120 hours after the Pringle maneuver, on the postoperative period. In all cases peripheral venous blood was sampled at a site where no contamination with any of the infused fluids could occur. Immediately after obtaining the sample, the blood was centrifuged, the plasma isolated, and frozen to -70 degrees centigrade. In all cases, two plasma tubes of each sample were individually labeled and stored.

IL 6 was quantified by means of the IL 6 EASIA (BIOSOURCE, Europe Belgium) based on oligoclonal antibodies coupled with monoclonal antibodies to various IL 6 epitopes. This method showed sensitivity for both low and standard IL 6 ranges. The results expressed in our manuscript represent the mean of both samples obtained for each time point of the study.

Statistical analysis Based on previous studies that reported a \> 100 picograms/mL difference among both groups with a SD \< 50 picograms/mL, we based our calculations on a predicted difference among both groups of 50 picograms/mL with a SD of 50 picograms/mL (17). Based on the fact that since sixteen patients in each group would allow rejection of the null hypothesis with an 80% confidence in the setting of a difference \>50 picograms/mL among both groups, a total of 36 patients were randomized. P\<0.05 was considered significant.